Digital Signal Processing and Digital-to-analog Converters for Wide-band Transmitters

In this thesis, the implementation methods of digital signal processing and digitalto-analog converters for wide-band transmitters are researched. With digital signal processing, the problems of analog signal processing, such as sensitivity to interference and nonidealities of the semiconductor processes, can be avoided. Also, the programmability can be implemented digitally more easily than by means of analog signal processing. During the past few years, wireless communications has evolved from analog to digital, and signal bandwidths have increased, enabling faster and faster data transmission. The evolution of semiconductor processes, decreasing linewidth and supply voltages, has decreased the size of the electronics and power dissipation, enabling the integration of larger and larger systems on single silicon chips. There is little overall benefit in decreasing linewidths to meet the needs of analog design, since it makes the design process more difficult as the device sizes cannot be scaled according to minimum linewidth and because of the decreasing supply voltage. On the other hand, the challenges of digital signal processing are related to the efficient realization of signal processing algorithms in such a way that the required area and power dissipation does not increase extensively. In this book, the problems related to digital filters, upconversion algorithms and digital-to-analog converters used in digital transmitters are researched. Research results are applied to the implementation of a transmitter for a third-generation WCDMA base-station. In addition, the theory of factors affecting the linearity and performance of digitalto analog converters is researched, and a digital calibration algorithm for enhancement of the static linearity has been presented. The algorithm has been implemented together with a 16-bit converter; its functionality has been demonstrated with measurements.